The process of assembly and secretion of triglyceride-rich lipoproteins (TG-LP) such as VLDL is a complicated process requiring lipids and the N- terminal 31-41% of apo-B (B). Current hypotheses suggest that there are 2 pathways: if lipid is available, B is secreted on nascent TG-rich particles - if lipid is deficient, B is degraded. In the secretory path a primordial HDL/LDL sized particle is first released into the ER, and then it is enlarged to form VLDL. Several possible pause sequences in the N-terminal domain might facilitate lipid binding. Mammary derived C127 cells make no apolipoproteins, secrete no lipid nor have microsomal TG transfer protein (MTP). When C127 cells are transfected with cDNA for C truncated B forms, they effectively secrete B17 in the lipid poor state but secrete B41 exclusively on primordial TG-rich particles (about 125 angstroms). Thus, B41 translated by C127 cells also follows two pathways: secretion or degradation. The information for the assembly with TG, detachment from the ER membrane and secretion, therefore resides in the sequences between apo-B17 and apo-B41. A multialgorithm search for potential lipid binding sequences in B41 indicates that there is a large region very rich in amphipathic beta sheets (ABS) located between B21 and B41. Thirty-five 12aa strands have been identified. There are 7 putative pause sequences also in this region and 5 overlap ABS. Constructs expressing truncated forms between apo-B29 and apo-B41 will be made to determine the minimum N-terminal required to commit B to formation of a primordial TG-rich particle. Mutations within key ABS domains and pause sequences in B21-B41 will be made to examine which B microstructures are necessary to permit association with TG to form particles. Consensus ABS sequences will be synthesized and combined with membrane lipids and TG, and this structure will be studied by X-ray, NMR, and cryo-electron microscopy. Consensus sequences will be used to design oligonucleotides which link 10 to 40 consensus strands through beta turns. These will be ligated to B17 and B29 to produce chimeric proteins potentially capable of forming primordial TG-LP. Chaperones which bind to B during the assembly process will be probed. The 3D structure of the secreted particles and ER lumenal particles will be explored by cryo-electronmicroscopy. Our hypothesis is that pause sequences allow ABS to intercalate into the ER membrane region of the translocon. If TG is adjacent, ABS will bind and initiate formation of the core of a primordial particle. If lipid is not found, then the protein acting as a foreign transmembrane protein, is degraded.